Dissemin is shutting down on January 1st, 2025

Published in

IOP Publishing, Environmental Research Letters, 3(4), p. 035007, 2009

DOI: 10.1088/1748-9326/4/3/035007

Links

Tools

Export citation

Search in Google Scholar

Influence of soil heterogeneity on evapotranspiration under shallow water table conditions: Transient stochastic simulations

Journal article published in 2009 by Stefan J. Kollet ORCID
This paper is made freely available by the publisher.
This paper is made freely available by the publisher.

Full text: Download

Green circle
Preprint: archiving allowed
Green circle
Postprint: archiving allowed
Green circle
Published version: archiving allowed
Data provided by SHERPA/RoMEO

Abstract

Ensembles of soil column numerical experiments were performed to study the influence of heterogeneity in the saturated hydraulic conductivity on the evapotranspiration or latent heat, LE, under varying water table conditions. In the numerical experiments, a variably saturated groundwater flow model coupled with a land surface model (ParFlow[CLM]) was used. The model was forced at the top with an atmospheric time series from Oklahoma, USA. The heterogeneity was simulated for two different soils (clay and sand) using uncorrelated Gaussian random fields with different variances. Soil heterogeneity has a strong influence on LE during the dry months of the year and negligible influence during months with sufficient moisture availability. The influence is stronger for unstructured soils, such as sand. An increase in shallow water table depth collapses the ensemble onto a single curve, i.e., heterogeneity plays a minor role and LE is limited by the increasing redistribution distance from the water table to the land surface. Accounting for correlations between the hydraulic conductivity and the shape factor in the pressure–saturation function increases the variability in LE. In the case of laterally interconnected soil columns, 3D hydrodynamics homogenize LE fluxes by establishing additional flow paths toward the land surface. Comparison with geometric mean simulations shows good agreement over many time periods for weekly and monthly averaged LE fluxes. Instantaneous daily and hourly differences sometimes exhibit values on the order of 100–101 W m−2 (10−2–10−1 mm d−1) and white noise behavior over extended time periods. In the simulations, perhaps the strongest limitation is the application of a constant atmospheric time series at the top of the simulation domain. In order to assess the impact of this limitation in simulations of subsurface–land surface–atmosphere interactions, it would be necessary to allow for two-way feedbacks with the lower atmosphere. This will require resolving the boundary layer immediately adjacent to the land surface, including the lateral exchange of mass, energy, and momentum.